Econ 522
Economics of Law
Dan Quint
Spring 2011
Lecture 18
So far…
 We’ve discussed a bunch of liability rules…



No liability
Strict liability
Various versions of a negligence rule
 …and the effect of each rule on incentives for:




Injurer precaution
Victim precaution
Injurer activity level
Victim activity level
1
So far…
 And we discussed effect of errors in calculating damages
or standard for negligence
 Strict liability rule


random errors in calculating damages have no effect
systematic errors effect injurer behavior
 Negligence rule



small errors (random or systematic) in calculating damages have no
effect on injurer precaution
errors in setting legal standard for negligence have strong effect on
injurer precaution
uncertainty in legal standard leads to overprecaution
2
Up next…
 What factors/complications has our simple model been
leaving out?
 How much money is your life worth?
 But first…
3
Experiment
4
Relaxing the
assumptions
of our model
5
Our model thus far has assumed…
 So far, our model has assumed:

People are rational

Injurers pay damages in full

They don’t run out of money and go bankrupt

There are no regulations in place other than the liability rule

There is no insurance

Litigation is costless
 We can think about what would happen when each of
these assumptions is violated
6
Assumption 1: Rationality
 Behavioral economics: people systematically misjudge
value of probabilistic events
 Daniel Kahneman and Amos Tversky, “Prospect Theory: An
Analysis of Decision under Risk”







45% chance of $6,000 versus 90% chance of $3,000
Most people (86%) chose the second
0.1% chance of $6,000 versus 0.2% chance of $3,000
Most people (73%) chose the first
But under expected utility, either u(6000) > 2 u(3000), or it’s not
So people don’t actually seem to be maximizing expected utility
And the “errors” have to do with how people evaluate probabilities
7
Assumption 1: Rationality
 People seem to overestimate chance of unlikely events with
well-publicized, catastrophic events
 Freakonomics: people fixate on exotic, unlikely risks, rather
than more commonplace ones that are more dangerous
8
Assumption 1: Rationality
 People seem to overestimate chance of unlikely events with
well-publicized, catastrophic events
 Freakonomics: people fixate on exotic, unlikely risks, rather
than more commonplace ones that are more dangerous
 How to apply this: accidents with power tools





Could be designed safer, could be used more cautiously
Suppose consumers underestimate risk of an accident
Negligence with defense of contributory negligence: would lead to
tools which are very safe when used correctly
But would lead to too many accidents when consumers are irrational
Strict liability would lead to products which were less likely to cause
9
accidents even when used recklessly
Assumption 1: Rationality
 Another type of irrationality: unintended lapses
 “Many accidents result from tangled feet, quavering hands,
distracted eyes, slips of the tongue, wandering minds,
weak wills, emotional outbursts, misjudged distances, or
miscalculated consequences”
10
Assumption 2: Injurers pay damages in full
 Strict liability: injurer internalizes expected harm done,
leading to efficient precaution
 But what if…






Harm done is $1,000,000
Injurer only has $100,000
So injurer can only pay $100,000
But if he anticipates this, he knows D << A…
…so he doesn’t internalize full cost of harm…
…so he takes inefficiently little precaution
 Injurer whose liability is limited by bankruptcy is called
judgment-proof
11
Example of judgment-proofness
(from old final exam)
 Owner of an oil tanker
 Any accident would be an environmental catastrophe,
doing $50,000,000 of harm
 Upgraded navigation system would cost $225,000, and
reduce likelihood of an accident from 1/100 to 1/500


Precaution reduces expected harm from $500,000 to $100,000,
costs $225,000, so efficient to take precaution
If company would be forced to pay $50,000,000 after an accident,
then under strict liability, would choose to buy new nav system
 Suppose the business is only worth $5,000,000



If there’s an accident, pay the $5,000,000 and go out of business
Now nav system reduces expected damages from $50,000 to
$10,000 – not worth the cost
12
So judgment-proof business would take too little precaution
Assumption 3: No regulation
 What stops me from speeding?


If I cause an accident, I’ll have to pay for it
Even if I don’t cause an accident, I might get a speeding ticket
 Similarly, fire regulations might require a store to have a
working fire extinguisher
 When regulations exist, court could use these standards as
legal standard of care for avoiding negligence

Or court might decide on a separate standard
13
Continuing the example of judgment-proofness
from before…
 We saw, if business is only worth $5,000,000, liability does
not create enough incentive to upgrade nav system
 Now suppose government passes regulation requiring
modern navigation systems on all oil tankers

If business doesn’t upgrade, 1 in 5 chance of being caught by
safety inspector and having to pay a $1,000,000 fine
 Now, combining liability with regulation…



Upgrade: cost of new nav system is $225,000, expected damages
are $10,000  private cost is $235,000
Don’t upgrade: expected damages are $50,000, expected
government fine is $200,000  private cost is $250,000
Liability + regulation gives enough incentive to take precaution,
even though either one alone would not be enough
14
Assumption 3: No regulation
 When liability > injurer’s wealth, liability does not create
enough incentive for efficient precaution
 Regulations which require efficient precaution solve the
problem
 Regulations also work better than liability when accidents
impose small harm on large group of people
15
Assumption 4: No insurance
 We assumed injurer or victim actually bears cost of
accident
 When injurer or victim has insurance, they no longer have
incentive to take precaution
 But, insurance tends not to be complete
16
Assumption 4: No insurance
 Insurance reduces incentive to take precaution

Moral hazard
 Insurance companies have ways to reduce moral hazard



Deductibles, copayments
Increasing premiums after accidents
Insurers may impose safety standards that policyholders must meet
17
Assumption 5: Litigation costs nothing
 If litigation is costly, this affects incentives in both directions




If lawsuits are costly for victims, they may bring fewer suits
Some accidents “unpunished”  less incentive for precaution
But if being sued is costly for injurers, they internalize more than the
cost of the accident
So more incentive for precaution
 A clever (unrealistic) way to reduce litigation costs





At the start of every lawsuit, flip a coin
Heads: lawsuit proceeds, damages are doubled
Tails: lawsuit immediately dismissed
Expected damages are the same  same incentives for precaution
But half as many lawsuits to deal with!
18
More twists
on liability
19
Vicarious Liability
 Vicarious liability is when one person is held liable
for harm caused by another


Parents may be liable for harm caused by their child
Employer may be liable for harm caused by employee
 Respondeat superior – “let the master answer”
 Employer is liable for unintentional torts of employee
if employee was acting within the scope of his
employment
20
Vicarious Liability
 Gives employers incentive to...



be more careful who they hire
be more careful what they assign employees to do
supervise employees more carefully
 Employers may be better able to make these decisions
than employees…
 …and employees may be judgment-proof
21
Vicarious Liability
 Vicarious liability can be implemented through…


Strict liability rule: employer liable for any harm caused by
employee (as long as employee was acting within scope of
employment)
Negligence rule: employer is only liable if he was negligent in
supervising employee
 Which is better? It depends.


If proving negligent supervision is too hard, strict vicarious liability
might work better
But an example favoring negligent vicarious liability…
22
Joint and Several Liability
 Suppose you were harmed by accident caused by two
injurers
 Joint liability: you can sue them both together
 Several liability: you can sue each one separately

Several liability with contribution: each is only liable for his share
of damage
 Joint and several liability: you can sue either one for the
full amount of the harm

Joint and several liability with contribution: the one you sued
could then sue his friend to get back half his money
23
Joint and Several Liability
 Joint and several liability holds under common law
when…


Defendants acted together to cause the harm, or…
Harm was indivisible (impossible to tell who was at fault)
 Good for the victim, because…


No need to prove exactly who caused harm
Greater chance of collecting full level of damages

Instead of suing person most responsible, could sue person most likely
to be able to pay
24
Back to Comparative Negligence
 Negligence with a defense of contributory negligence
was dominant liability rule in common law countries



Negligent injurer is liable, unless victim was also negligent
Example: a car going 60 mph hits a car going 35 in a 30-mph zone
Since victim was also negligent, injurer is not liable
 Last 40 years, most U.S. states have adopted a
comparative negligence rule




Usually through legislation, sometimes through judicial decision
Appealing from fairness point of view
But any negligence rule leads to efficient precaution
So how do we explain the move?
25
Comparative Negligence and Evidentiary
Uncertainty
 Evidentiary uncertainty



Given a legal standard for negligence, xn…
…and an actual level of precaution taken, x…
still uncertainty in whether the court will find negligence
 Evidentiary uncertainty, like random errors in setting xn,
leads to over-precaution…
 …but comparative negligence partly mitigates this
26
Simple negligence,
evidentiary uncertainty
$
Comparative negligence,
evidentiary uncertainty
Any negligence rule
Comparative negligence and evidentiary
uncertainty
wx + p(x) A
wx
p(x) A
x*
x
 Comparative negligence mitigates effect of evidentiary
uncertainty
27
Perfect
Compensation
28
Perfect compensation
 Perfect compensatory damages (D = A)



Returns victim to original level of well-being
(Works like insurance)
And sets correct incentive for injurers
 But in some cases, hard to determine level


Might be no price at which you’d be willing to give up a leg
Certainly no price at which a parent would be indifferent toward
losing a child
29
Perfect compensation
 Recommended jury instructions, Massachusetts:




“Recovery for wrongful death represents damages to the survivors
for the loss of value of decedent’s life. There is no special formula
under the law to assess the plaintiff’s damages…
It is your obligation to assess what is fair, adequate, and just.
You must use your wisdom and judgment and your sense of basic
justice to translate into dollars and cents the amount which will fully,
fairly, and reasonably compensate the next of kin for the death of the
decedent.
You must be guided by your common sense and your conscience
on the evidence of the case…”
 And from California:

“…You should award reasonable compensation for the loss of love,
companionship, comfort, affection, society, solace or moral support.”
30
One other odd feature of compensatory
damages…
 Most people would rather be horribly injured than killed
 Which means killing someone does more damage than
injuring someone
 But compensatory damages tend to be lower for a fatal
accident than an accident which crippled someone



When someone is badly injured, may require huge amount of
money to compensate them
In wrongful-death case, damages compensate victim’s loved ones,
but no attempt to compensate victim
So these damages tend to be smaller
31
What’s a life
worth?
32
What’s a life worth?
 Assessing damages in a wrongful death lawsuit requires
some notion of what a life is worth
 Safety regulators also need some notion of what a life is
worth
 Kip Viscusi, The Value of Risks to Life and Health
Regulation
Estimated cost per life saved
Airplane cabin fire protection
$
200,000
Car side door protection standards
$
1,300,000
OSHA asbestos regulations
$
89,300,000
EPA asbestos regulations
$
104,200,000
Proposed OSHA formaldehyde standard
$72,000,000,000
 Regulators need to decide “where to draw the line”
33
Kip Viscusi, The Value of Risks to Life and
Health
 Let w be starting wealth, D death, p probability
 There might be some amount of money M such that
p u(D) + (1 – p) u(w + M) = u(w)





When p = 1, this breaks down not because you can’t equate death
with compensation, but because the second term vanishes
So how do we find M?
Ask a bunch of people how much money they would need to take a
1/1000 chance of death?
Can’t do a lab experiment where you actually expose people to a
risk of death!
Clever trick: impute how much compensation people require from
the real-life choices they make
34
Kip Viscusi, The Value of Risks to Life and
Health
 Lots of day-to-day choices increase or decrease our risk of
death



Choose between Volvo and sports car with fiberglass body
Take a job washing skyscraper windows, or office job that pays less
Buy smoke detectors and fire extinguishers, or don’t
 “Hand Rule Damages”




Hand Rule: precaution is cost-justified if
cost of precaution < reduction in accidents X cost of accident
Suppose side-curtain airbags reduce risk of fatal accident by 1/1000
If someone pays $1,000 extra for a car with side-curtain airbags, it
must mean that
$1,000 < 1/1000 * value of their life
or, they value their life more than $1,000,000
35
Kip Viscusi, The Value of Risks to Life and
Health
 Viscusi surveys lots of existing studies which impute value
of life from peoples’ decisions
 Many use wage differentials

How much higher are wages for risky jobs compared to safe jobs?
 Others look at…



Decisions to speed, wear seatbelts, buy smoke detectors, smoke
cigarettes
Decision to live in very polluted areas (comparing property values)
Prices of newer, safer cars versus older, more dangerous ones
 Some used surveys to ask how people would make tradeoffs
between money and safety
 Each paper reaches some estimate for implicit value people
attach to their lives
36
What does Viscusi find?
37
24 studies based on wage differentials
20,000,000
15,000,000
Implicit
value
of life
10,000,000
5,000,000
0
3)
7) 4)
6)
1) 8)
1)
1)
1) 89
2)
8)
4)
5)
5)
3)
8) 0)
1)
0)
4)
6)
4)
8)
8)
0)
7)
9)
1)
99 98 197 198 198 198 198 199 98 198 199 198 198 97 197 198 98 197 98 98 199 (19 198 198 198 198 198 99 99
1
1
1
(
1
1
(
(
(
(1 t (1
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(1 i (1
(
,
r
.
l
e
i
i
i
i
s
l
i
n
n
h
h
h
s
l
n
n
r
e
h
h
n
8
r
h
a
t
s
s
t
s
a
it
et lbe ose ho ham utl ow t. a cus ulo eet et
om om ig
am 7
re cus cus
so cu mi
oo m
cu cu Lee
i
e
B Br
ic
Le
L
s
g
gh (19 Sm Ol
is
M lott Fols Fols Le Ga Vis Vis
S Vis Vis
u
po
Vi
& & G & R & N illin
& i&
h &
ax & V ars r & nea illin usi
r
r
c
&
&
&
&
&
g
r
e ith
e
e
s
i
D
D sc
e
h
e
ld
re
re sn cu & S igh igh
re
re
G ore ac esn ous
sn
al
Vi
ie Sm Th nou
i
e Le
o Ps
oo oo nie Vis og
oo oo
C
n
n
L
r
M &
M
M K
M
M
K
K
z
A
er
in
H
ar
M
38
7 studies using other risk-money tradeoffs
Nature of Risk,
Year
Component of the
Monetary Tradeoff
Implicit Value of life
($ millions)
Highway speed-related
accident risk, 1973
Value of driver time based on
wage rates
0.07
Automobile death risks, 1972
Estimated disutility of seat
belts
1.2
Fire fatality risks without smoke
detectors, 1974-1979
Purchase price of smoke
detectors
0.6
Mortality effects of air pollution,
1978
Property values in Allegheny
Co., PA
0.8
Cigarette smoking risks, 1980
Estimated monetary
equivalent of effect of risk info
0.7
Fire fatality risks without smoke
detectors, 1968-1985
Purchase price of smoke
detector
2.0
Automobile accident risks,
1986
Prices of new automobiles
4.0
39
6 studies based on surveys
Nature of
Risk
Survey
Methodology
Implicit Value of Life
($ millions)
Improved ambulance service,
post-heart attack lives
Willingness to pay question, door-to-door
small (36) Boston sample
0.1
Airline safety and locational
life expectancy risks
Mail survey willingness to accept increased
risk, small (30) U.K. sample, 1975
15.6
Job fatality risk
Willingness to pay, willingness to accept
change in job risk in mail survey, 1984
3.4 (pay),
8.8 (accept)
Motor vehicle accidents
Willingness to pay for risk reduction, U.K.
survey, 1982
3.8
Automobile accident risks
Interactive computer program with pairwise
auto risk-living cost tradeoffs until
indifference achieved, 1987
2.7 (median)
9.7 (mean)
Traffic safety
Series of contingent valuation questions,
New Zealand survey, 1989-1990
1.2
40
Kip Viscusi, The Value of Risks to Life and
Health
 Wide range of results


Most suggest value of life between $1,000,000 and $10,000,000
Many clustered between $3,000,000 and $7,000,000
 Even with wide range, he argues this is very useful:





“In practice, value-of-life debates seldom focus on whether the
appropriate value of life should be $3 or $4 million…
However, the estimates do provide guidance as to whether risk
reduction efforts that cost $50,000 per life saved or $50 million per
life saved are warranted.”
“The threshold for the Office of Management and Budget to be
successful in rejecting proposed risk regulations has been in excess
of $100 million.”
C&U: NHTSA uses $2.5 million for value of traffic fatality
Current: EPA $9.1 MM, FDA $7.9 MM, Transpo Dept $6 MM
41
Punitive
damages
42
Inconsistency of damages
 Damage awards vary greatly across countries, even across
individual cases
 We saw last week:

As long as damages are correct on average, random inconsistency
doesn’t affect incentives (under either strict liability or negligence)
 But, if appropriate level of damages isn’t well-established,
more incentive to spend more fighting
43
Punitive damages
 What we’ve discussed so far: compensatory damages

Meant to “make victim whole”/compensate for actual damage done
 In addition, courts sometimes award punitive damages


Additional damages meant to punish injurer
Create stronger incentive to avoid initial harm
 Punitive damages generally not awarded for innocent
mistakes, but may be used when injurer’s behavior was
“malicious, oppressive, gross, willful and wanton, or
fraudulent”
44
Punitive damages
 Calculation of punitive damages even less well-defined
than compensatory damages
 Level of punitive damages supposed to bear “reasonable
relationship” to level of compensatory damages


Not clear exactly what this means
U.S. Supreme Court: punitive damages more than ten times
compensatory damages will attract “close scrutiny,” but not explicitly
ruled out
45
Example of punitive damages: Liebeck v
McDonalds (1994) (“the coffee cup case”)
 Stella Liebeck was badly burned when she spilled a cup of
McDonalds coffee in her lap
 Awarded $160,000 in compensatory damages, plus $2.9
million in punitive damages
 Case became “poster child” for excessive damages, but…
46
Liebeck v McDonalds (1994)
 Stella Liebeck dumped coffee in her lap while adding
cream/sugar



Third degree burns, 8 days in hospital, skin grafts, 2 years treatment
Initially sued for $20,000, mostly for medical costs
McDonalds offered to settle for $800
 McDonalds serves coffee at 180-190 degrees




At 180 degrees, coffee can cause a third-degree burn requiring skin
grafts in 12-15 seconds
Lower temperature would increase length of exposure necessary
McDonalds had received 700 prior complaints of burns, and had
settled with some of the victims
Quality control manager testified that 700 complaints, given how
many cups of coffee McDonalds serves, was not sufficient for
McDonalds to reexamine practices
47
Liebeck v McDonalds (1994)
 Rule in place was comparative negligence






Jury found both parties negligent, McDonalds 80% responsible
Calculated compensatory damages of $200,000
times 80% gives $160,000
Added $2.9 million in punitive damages
Judge reduced punitive damages to 3X compensatory, making total
damages $640,000
During appeal, parties settled out of court for some smaller amount
 Jury seemed to be using punitive damages to punish
McDonalds for being arrogant and uncaring
48
What is the economic purpose of punitive
damages?
 We’ve said all along: with perfect compensation, incentives
for injurer are set correctly. So why punitive damages?
 Example…







Suppose manufacturer can eliminate 10 accidents a year, each
causing $1,000 in damages, for $9,000
Clearly efficient
If every accident victim would sue and win, company has incentive to
take this precaution
But if some won’t, then not enough incentive
Suppose only half the victims will bring successful lawsuits
Compensatory damages would be $5,000; company is better off
paying that then taking efficient precaution
One way to fix this: award higher damages in the cases that are
brought
49
This suggests…
 Punitive damages should be related to compensatory
damages, but higher the more likely injurer is to “get away
with it”




If 50% of accidents will lead to successful lawsuits, total damages
should be 2 X harm
Which requires punitive damages = compensatory damages
If 10% of accidents lead to awards, damages should be 10 X harm
So punitive damages should be 9 X compensatory damages
 Seems most appropriate when injurer’s actions were
deliberately fraudulent, since may have been based on costbenefit analysis of chance of being caught
50
Some empirical observations
about tort system in the U.S.
(won’t get to this)
51
U.S. tort system
 In 1990s, tort cases passed contract cases as most
common form of lawsuit


Most handled at state level: in 1994, 41,000 tort cases resolved in
federal courts, 378,000 in state courts in largest 75 counties
Most involve a single plaintiff (many contract cases involve multiple
plaintiffs)
 Among tort cases in 75 largest U.S. counties…




60% were auto accidents
17% were “premises liability” (slip-and-fall in restaurants,
businesses, government offices, etc.)
5% were medical malpractice
3% were product liability
52
U.S. tort system
 Punitive damages historically very rare



1965-1990, punitive damages in product liability cases were
awarded 353 times
Average damage award was $625,000, reduced to $135,000 on
appeal
Average punitive damages only slightly higher than compensatory
 In many states, punitive damages limited, or require higher
standard of evidence


Civil suits generally require “preponderance of evidence”
In many states, punitive damages require “clear and convincing”
evidence
53
U.S. tort system
 Medical malpractice



New York study in 1980s: 1% of hospital admissions involved
serious injury due to negligent care
Some estimates: 5% of total health care costs are “defensive
medicine” – procedures undertaken purely to prevent lawsuits
Some states have considered caps on damages for medical
malpractice
54
U.S. tort system
 Product liability

Recent survey of CEOs: “liability concerns caused 47% of those
surveyed to drop one or more product lines, 25% to stop some
research and development, and 39% to cancel plans for a new
product.”
 Liability standard for product-related accidents is “strict
products liability”




Manufacturer is liable if product determined to be defective
Defect in design
Defect in manufacture
Defect in warning
55
Vaccines
 Most vaccines are weakened version of disease itself



Make you much less likely to acquire the disease
But often come with very small chance of contracting disease
directly from vaccine
Salk polio vaccine wiped out polio, but caused 1 in 4,000,000
people vaccinated to contract polio
 1974 case established maker had to warn about risk



Since then, some people were awarded damages after their
children developed polio from vaccine
If liability can’t be avoided, built into cost of the drug
And discourages companies from developing vaccines
56
Mass torts
 Since health risks of asbestos understood, over 600,000
people have brought lawsuits against 6,000 defendants
 DES (drug administered to pregnant women in 1950s)


Impossible to establish which firm produced dose given to a
particular woman
California Supreme Court introduced “market share liability”
 Class action lawsuit




Small, dispersed harms – no plaintiff might find it worthwhile to sue
Class action suits allow large lawsuits with lots of plaintiffs
Give more incentive for precaution against diffuse harms
But…
57
Cooter and Ulen’s overall assessment of
U.S. tort system
 Critics claim juries routinely hand out excessive awards
and tort system is out of control…
 …but actually it functions reasonably well
 Outside of occasional, well-publicized outliers, damage
awards are generally reasonable…
 …and liability has led to decreases in accidents in many
industries
58
To wrap up tort law, a funny story from
Friedman…
“A tort plaintiff succeeded in collecting a large damage judgment.
The defendant’s attorney, confident that the claimed injury was
bogus, went over to the plaintiff after the trial
and warned him that if he was ever seen out of his wheelchair
he would be back in court on a charge of fraud.
The plaintiff replied that to save the lawyer the cost of having
him followed, he would be happy to describe his travel plans.
He reached into his pocket and drew out an airline ticket –
to Lourdes, the site of a Catholic shrine famous for miracles.”
59